Changeset 1605

Timestamp:

07/09/11 08:46:46 (11 months ago)

Author:

jharrop

Message:

Proof of concept of an mc:AlternateContent preprocessor.

Location:

trunk/docx4j/src/main/java/org/docx4j

Files:

• jaxb/JaxbValidationEventHandler.java (modified) (3 diffs)
• jaxb/mc-preprocessor.xslt (added)
• openpackaging/parts/WordprocessingML/MainDocumentPart.java (modified) (3 diffs)
• utils/XSLTUtils.java (added)

trunk/docx4j/src/main/java/org/docx4j/jaxb/JaxbValidationEventHandler.java

@@ -21 +21 @@
 package org.docx4j.jaxb;
 
+import java.io.IOException;
+
 import javax.xml.bind.ValidationEvent;
 import javax.xml.bind.ValidationEventHandler;
 import javax.xml.bind.ValidationEventLocator;
+import javax.xml.transform.Source;
+import javax.xml.transform.Templates;
+import javax.xml.transform.TransformerConfigurationException;
+import javax.xml.transform.stream.StreamSource;
 
 import org.apache.log4j.Logger;
+import org.docx4j.XmlUtils;
 
 
@@ -33 +40 @@
         private static Logger log = Logger.getLogger(JaxbValidationEventHandler.class);         
         
+        private boolean shouldContinue = true;
+        public void setContinue(boolean val) {
+                shouldContinue = val;
+        }
+        
+        public final static String UNEXPECTED_MC_ALTERNATE_CONTENT = "unexpected element (uri:\"http://schemas.openxmlformats.org/markup-compatibility/2006\", local:\"AlternateContent\")";
+        
+        static Templates mcPreprocessorXslt;    
+        
+        public static Templates getMcPreprocessor() throws IOException, TransformerConfigurationException {
+                
+                if (mcPreprocessorXslt==null) {
+                        try {
+                                Source xsltSource  = new StreamSource(
+                                                org.docx4j.utils.ResourceUtils.getResource(
+                                                                "org/docx4j/jaxb/mc-preprocessor.xslt"));
+                                mcPreprocessorXslt = XmlUtils.getTransformerTemplate(xsltSource);
+                        } catch (IOException e) {
+                                log.error(e);
+                                throw(e);
+                        } catch (TransformerConfigurationException e) {
+                                log.error(e);
+                                throw(e);
+                        }
+                }
+                
+                return mcPreprocessorXslt;
+                
+        }
         
     public boolean handleEvent(ValidationEvent ve) {            
@@ -69 +105 @@
       // (Marshalling, Unmarshalling, Validating)
       log.info("continuing (with possible element/attribute loss)");
-       return true;
+       return shouldContinue;
              
      }

trunk/docx4j/src/main/java/org/docx4j/openpackaging/parts/WordprocessingML/MainDocumentPart.java

@@ -33 +33 @@
 import javax.xml.bind.JAXBElement;
 import javax.xml.bind.JAXBException;
+import javax.xml.bind.UnmarshalException;
 import javax.xml.bind.Unmarshaller;
+import javax.xml.bind.util.JAXBResult;
 import javax.xml.parsers.DocumentBuilderFactory;
+import javax.xml.transform.Result;
+import javax.xml.transform.Templates;
+import javax.xml.transform.dom.DOMResult;
 
 import org.apache.log4j.Logger;
@@ -42 +47 @@
 import org.docx4j.dml.CTNonVisualDrawingProps;
 import org.docx4j.jaxb.Context;
+import org.docx4j.jaxb.JaxbValidationEventHandler;
 import org.docx4j.model.PropertyResolver;
 import org.docx4j.model.listnumbering.AbstractListNumberingDefinition;
@@ -269 +275 @@
                         binder = jc.createBinder();
                         
-                        binder.setEventHandler(
-                                        new org.docx4j.jaxb.JaxbValidationEventHandler());
-                        
-                        jaxbElement =  (org.docx4j.wml.Document) binder.unmarshal( doc );
+                        JaxbValidationEventHandler eventHandler = new JaxbValidationEventHandler();
+                        eventHandler.setContinue(false);
+                        binder.setEventHandler(eventHandler);
+                        
+                        try {
+                                jaxbElement =  (org.docx4j.wml.Document) binder.unmarshal( doc );
+                        } catch (UnmarshalException ue) {
+
+                                // mimic docx4j 2.7.0 and earlier behaviour
+                                eventHandler.setContinue(true);
+                                
+                                if (ue.getMessage().contains(JaxbValidationEventHandler.UNEXPECTED_MC_ALTERNATE_CONTENT)) {
+                                        // Try our preprocessor
+                                        log.info("encountered mc:AlternateContent; pre-processing");
+                                        
+                                        // There is no JAXBResult(binder),
+                                        // so use a 
+                                        DOMResult result = new DOMResult();
+                                        
+                                        Templates mcPreprocessorXslt = JaxbValidationEventHandler.getMcPreprocessor();
+                                        XmlUtils.transform(doc, mcPreprocessorXslt, null, result);
+                                        
+                                        doc = (org.w3c.dom.Document)result.getNode();
+                                        
+                                } else {
+                                        log.error(ue);
+                                        log.info("trying again; likely attribute/element loss");                                        
+                                }
+                                jaxbElement =  (org.docx4j.wml.Document) binder.unmarshal( doc );                                       
+                                
+                        }
                         
                         return jaxbElement;